Well I have printed out the full size D800 samples that are available. I know the current thinking in DSLR land is that you can never have too much detail.
Yeah, you can tell what fantastic detail there is just viewing these jpgs but to me, the real proof is in the print. Just as expected, the detail in an actual print is just as astounding as you would think it would be and and while it is amazing, I also see that it may require even more percise post processing when it comes to close up face shots.
Take for example the bride with the veil. The detail in the veil is amazing but this same crisp detail brings out the flaws in the bride's skin, and I can see it bringing out flaws in makeup as well. I have had to deal with flaws in makeup with my little 12MP D2X. On the below print you cannot really see the contour of the blemishes under the makeup but in the full size print, it is pretty glaring. Of course in a lessor MP camera, these flaws are somewhat diffused when increasing the size of a print but when you already start out with a native resolution as high as D800 file, then you don't really benefit the subtle softness that comes with upsizing.
Now this brings up yet another challenge. I suspect post processing will take more skill as well when it comes to retouching. Where flaws in retouching may be diffused when upsizing a print, when the file is already as big as the 36MP file, then any flaws in post processing will be more apparent. Also, overprocessing will be more noticeable.
In summary, the fantastic detail that all are amazed over that is in these 36MP files can be a blessing and a curse depending on what you are shooting. This is not to diss the D800 in any way. This is not some lame justification why I will not buy the D800 because I do plan to buy it.
The purpose of this post is to point out something I haven't seen posted in the gazillion of posts but then again, I haven't really read most of them that closely. This just popped out at me looking at actual prints made by these files. If this aspect has already been beaten to death, then my apologies.
Generally speaking, in post work on portraits where natural skin flaws are distracting, it's very easy and quick to "take the edge off" detail in complexion, while leaving all the other parts of the image razor sharp where, and as, desired, such as eyes, mouth, hair, jewelry, clothing, etc.
And by "take the edge off," I'm not talking about those PS add-ons advertised with images turning a model with zits into some kind of plastic kewpie doll. UUUGGGHHH!!
When I am talking to a young woman (or anybody else, for that matter), I'm looking in her eyes. Maybe a glance to the mouth, a normal distraction during conversation. I'm not busy examining the texture of her skin.
Unfortunately, in a photograph, particularly a large size image, it's normal to wander all over the image, soaking up all the detail, desirable or not.
My philosophy is to try to recreate the image from a "conversation," not to put on display in glaring detail, all the natural, human artifacts of life. There is a degree to which this kind of post processing is "right," and points where it seems too much or too little to me.
And I might add, the crisp sharpness in the remainder of the image leaves an overall impression of a critically sharp, high quality photograph.
I'll take the 36MP; I'll take all the advantages, and deal with the disadvantages in a proper, and sensitive way.
Well the standard today seems to be that you make everyone look like a model worthy of a magazine cover, trim off excess weight while perhaps enhancing other body features. I am not into transforming people into something they aren't so that is why I routinely say "no" to requests of senior portraits, weddings, etc. Sure, I will do some retouching to smooth skin, etc but I do not want to put up with expectations of total transformations.
With regards to going to 36MP, I think one will have to relearn their retouching skills in the same manner that one may have to look at their shooting technique.
When models become aware of the razor sharp resolution of the D800, they may ask NOT to be photographed with it. I think the bride looks HARD. Nikon made a point to tell us to "open up" the aperture to make portraits look SOFT. (Of course this was only suggested so we would thereby be using a faster shutter speed to combat the tendency for the D800 to produce BLUR.) I think the model looks HARD. I was surprised that Nikon didn't recommend the use of electronic flash to effectively increase the sharpness due to the short time duration of the flash pulse. Does electronic flash make models look even HARDER? I know that this depends on the "softness" of the light created by the photographer, but how practical is soft light when shooting a bride? She moves around. You move around. You don't always have a white wall to bounce the light from your flash. I doubt that the diffuser that comes with the flash will be soft enough.
I can hear it all now: Photographer to Model, "I just got a new D800 camera!" Model to Photographer, "Don't point that thing at me!"
I saw some really nice shots of my grandaughter taken by Olin Mills. Naturally, I wondered about the camera they used. It WAS A 5 MEGAPIXEL CAMERA!!!!!
I have some serious misgivings about EVER using my new D800E for portrature. Really, who would be that cruel? MY D700 IS ALREADY BORDERLINE FOR USE IN PORTRATURE. I won't be selling my D700. I need an FX camera for people pictures, AND THE D800E IS, IN MY HUMBLE OPINION, CRUEL OVERKILL.
I suppose it all depends on how much time and effort was spent in post-processing D800 images. So, as long as sufficient time is spent in post-processing, and the model finally looks good, there won't be any complaints. But woe to the photogragher who spends insufficient time, or simply doesn't have the skills himself or doesn't have the money to hire someone to do the post-processing for him so that the model will look good. That photographer will be shunned by all the models. Models are people. People talk.
And, I think that anyone who can afford a medium format digital camera, can probably afford to hire someone to do all the post-processing. The D800 is going to require a lot of post-processing if it is used for portrature. The question for me becomes, do I have the time to spend, or alternatively, do I want to spend the time with all this post-processing, or would I rather take portraits with my D700 which requires comparatively little post-processing time? I think I will stick to the D700 (I will not be selling it) for portrature, and use the D800E for something like landscape photography, where it should excel.
I don't know how relevant this is since it is a small file. Small files do hide a LOT whether it is poor skin or a poor application of makeup. It is not as if you could see the flaws that you see in the full size file to begin with. Why don't you download the full size file and run your routine? Some of the full size images I have seen, I am inclined to think one may have to start touching up on hands and fingers as well if you print the file without downsizing.
This is actually a very interesting thought. I saw your post on the other thread where you talked about the mascara flaw and that the fine facial hairs stand out. That's a really good point. The increased acuity of the D800 may force upon people the need to change their post processing routines a good deal. Of course I imagine that would be for prints of a certain size. I suspect many people will simply downsample images to fit the output medium.
Still, this may be signaling something for down the road when even more MP is available across the board.
I think it can actually be easier to retouch a larger photo. You have a bit more control over items that you want to clone or remove, and it's easier to isolate a change to just one area. Keep in mind that studio photographers doing similar work frequently use even higher megapixel medium format cameras that lack AA filters, and they get gorgeous results.
Thu 23-Feb-12 01:42 AM | edited Thu 23-Feb-12 01:53 AM by MelT
I am not too sure about bigger being easier Rick and I rarely disagree with you. Of course it depends on the ultimate size for the end result. I found playing around with the 16ish x 24ish native size sample file it was rather tough going because there is so much detail to be retouch. Think of fine facial hair that is VERY apparant in the shot above when viewed at full resolution. Even the detail in hands may need some work as well. Haven't seen adwork yet that celebrates the fine facial hair that may be on a woman .
Yes...medium format shooters need to deal with this but then again, how many photographers doing ad work, etc are actually doing their own processing? I would think instead they are simply turning over the files to the graphic artist to deal with. A professional graphic artist should have better skills doing this type of work with the photographer (or creative director) giving guidance what he/she wants.
I would think if you are at that level, a photographer would have someone on his/her staff to deal with the processing just like many years ago, we had a woman retouch negatives in the studio I worked for and the owner had me deal with the darkroom work but of course, he would approve or disapprove the final results or have me do minor tweaks.
I know if I was at that level of photography, I would have someone on staff to deal with the processing or outsource it to a graphic artist that I have developed a close relationship with who eventually learns my style, etc. Then again, I am not at the level so I am left to deal with these files with an overabundance of detail myself .
I think it's one of those things you just have to experience for yourself. I've been shooting portraits with a D3X for several years now and while going from 24MP to 36MP seems like a lot more, it's only slightly more than 20% more pixels in the linear sense. Not too big of deal. Portraits made with the D3X are easy to retouch because I can stay zoomed out a bit and work with larger areas (arms or cheeks) or zoom to have fine control over the eyes. If you take advantage of the many plug-ins out there for retouching skin (Nik, OnOne, etc.), you can get through the operations quickly, especially if you've done light retouching on your original image using LR, ACR, NX, etc. The bottom-line: it just hasn't been that hard, and I can't see another ~20% pixels breaking things. Also, my quick retouching operations have also been sanctioned by my wife, and she likes the portraits I've made of her. That should say a lot.
Also keep in mind that lighting can either enhance or supress fine detail in the skin. This specific shot was made to have the viewer think "holy moly, that's a lot of detail". You may not see as much in other shots depending on how you set your lights.
I hear you Rick. The way I see it, if you have a 12MP file blown to 16ish x 24ish and compare it to a 36MP file at native resolution of the same face, the details would be quite more distinct on the 36MP file. Would you agree?
Now my neophyte mind says that more aggressive means for smoothing the skin on the 36MP file would need to be taken regardless of choice of method (gaussian blur or a plug-in like Portraiture, etc.). On the 12MP file that is capturing less detail to begin with. This detail that is captured is softened to a degree when blowing up.
I also think that retouching done on a 12MP is softened when blown larger as well but with a 36MP file, you are starting off with a lot of troublesome detail.
Now I am not a big portrait guy at all but I have been able to retouch without going overboard whether it is skin defects or the misapplication of makeup. I just seeing both being greatly magnified on a 36MP file. No doubt, it can be dealt with but as I orginally wrote, I think the way one processes will change. I am scared to death of the peach fuzz I saw in the image shown and how I would deal with that. The peach fuzz I get at times is not as clear and sharp when shooting at 12MPs . I am now investigating how to deal with peach fuzz! I NEED A PEACH FUZZ PLUG-IN.
>I think it's one of those things you just have to experience >for yourself. I've been shooting portraits with a D3X for >several years now and while going from 24MP to 36MP seems like >a lot more, it's only slightly more than 20% more pixels in >the linear sense. Not too big of deal. ....
Are you keeping the D3x Rick - or trading it for a D800/800E?
Quite true but for those of us amateurs I think we might see more variation in output from those who can dedicate the time for processing (and have financial incentive) and those who can't.
The D800 is tempting for me but honestly I'd probably go with a D4. It will be a little more forgiving and still have great performance, maybe even better performance minus resolution. Of course, I doubt the D4 will actually improve my photography by all that much! Which is why I don't know if I will get one or not.
I think Thom Hogan's point for some time now (i.e. prior to D800 announcement) is capture optimum pixels, for maximum flexibility in post and as a hedge for the future when their use may be even more important. This model/veil photo happens to be the first high res one I looked at on the Nikon site and I was immediately taken with the amount of detail there (and yes I could see the peach fuzz!). There is a second photo of a black model in this series and the amount of detail there is also stunning!
I do somewhat routinely use a Capture NX2 Selection Control Point, Noise Reduction, and selective exclusion of eyes and lips already even at 12mp so yes this camera will require that also. Likewise I have always heard the more pixels the better when doing more intensive edits in post. I like the idea of using Gaussian Blur (rather than NR), I'll have to give that a try.
Good luck with your D800 I think it will be an amazing camera!
Steve...I have not read what Thom has to say on this but I do agree with the notion that maximum pixels is always better because you do not know what may happen in the future.
For me, there would definitely be a learning stage when dealing with the insane detail of the D800 files. Heck....going from a 4MP D2H to a 12MP D2X took some adjustment because detail was coming out on faces that weren't seen before . I will just have to keep women away from seeing there images before I have processed of I will have some hatin' women on my hands .
Fri 24-Feb-12 03:24 AM | edited Fri 24-Feb-12 03:32 AM by KnightPhoto
I use the noise reduction in 'File/Adjust' since I can apply it locally to skin using a selection control point. Be careful that you may have to unselect the NR option in 'File/Adjust' to start with if you had NR turned on in camera.
The NR available via 'Camera Settings' is not very useful since you can only apply it globally and I generally abhor the loss of detail that NR causes.
So what I do is: - I click once on the 'Selection Control Point' in the top bar of Capture NX2. Next I click on the tone I want to noise reduce (in this case I select skin, perhaps on the face). - Immediately the selected tone glows green to show the extent of the local tone selection that the adjustment will be applied to. - I can propagate the green selected tone to other areas of the image by holding the Option key (on my Mac) down and select the control point. This creates a second control point which I can drag to another area of the image that wasn't selected by the first control point I placed. (this step is not always necessary) - You can adjust the slider on the control point to the left to increase opacity (decrease strength) that the adjustment will be applied.
- Next in the active Edit Step on the Right-nav you can 'Select Adjustment' 'Noise Reduction...'. - Now I'll toggle "Faster" over to 'Better Quality'. - and for skin on a portrait I'll set 'Intensity' between 8-20 depending on age of subject - I'll reduce 'Sharpness' to 1 or 2.
- Finally I'll click once on the top Capture NX bar to 'Selection Brush' choosing the 'Minus' option. - I might then adjust the 'Size' of the brush with around 50 or less being a typical value. - I'll then brush the image by clicking once and holding, I'll then brush the Eyes. I'll let go and do a 2nd, 3rd, and 4th click/brush operation on lips, maybe eyebrows and ears too to keep them sharp by removing the noise reduction.
Takes me a lot longer to write this out than actually do it!
BTW I feel this model's beauty is enhanced by the wonderful clarity of the D800, I am very impressed with this camera.
Sat 25-Feb-12 02:19 PM | edited Sat 25-Feb-12 02:38 PM by Gromit44
Thanks for the detailed write-up Steve - I've tended to use the Selection Brush rather than the Selection Control Point but I'll give your method a try.
I agree the global NR in the 'Camera Settings' section is fairly useless, in fact I can't see why it's there at all. For a start, it's not actually a camera setting whereas all the others in that section are (i.e. WB, Picture Control & Active DL). Also, there's no point in having two NR options - they should just have added a 'Global on/off' switch to the NR in the Adjust section.
Well spotted Steve, you're right! I just checked two D700 shots with High ISO NR on and off.
If High ISO NR is switched on NX2's Camera Settings NR defaults to Intensity 22/Sharpness 7. If High ISO NR is switched off it defaults to Intensity 0/Sharpness 5. Maybe it interacts the same way with Long Exposure NR.
Fri 24-Feb-12 03:38 AM | edited Fri 24-Feb-12 03:39 AM by avisys
All this zits talk reminds me. I just looked at my B&W High School graduation picture (1956). I knew a studio photographer and retouching artist, so I saw what was done.
The photos were done with an 8x10 camera. The retoucher went to work with super-fine watercolor brushes, using cards of Kodak retouching media. She worked like a whiz 'cause she had to do every zit for every image for every student, even though she knew only one image would eventually be selected for each student. She was a true artist.
My photo looked like I had never had a zit, but I know I was covered with them.
I wonder what she would do with a D800 and Photoshop including some plug-ins.
Oh, the negative went into a huge 8x10 enlarger so it could be ~reduced~ to 4x5 or even a group of 2x3's.
How many effective pixels does an 8x10 negative have, compared to a D800???
The example image when viewed in its full size...a monster to download, I was a little critical of the light but the detail was spectacular and gave a sense of realism because of the small defects(which are not "defect" but are what was) like the facial fuzz greatly enhanced the image. While others find this to be an issue with being too real, I like it. Models will NOT object, it is a job or it is their event depending on whether they are modes or paying subjects. Pro models have more self confidence than some are assuming, they have to have, to subject themselves to such close scrutiny daily. It is the 30-50 year old paying client who will object but they do now with iPhone shots. I prefer shooting young people and old people, both are more comfortable with themselves. I do not shoot professionally but take a lot of portraits and candids that end up printed and on people's walls and found that being sensitive to their body image and feelings is part my job. When I like the subject and connect with them, the results show....which is why I have no interest in charging...its a hobby and I would prefer to work with people who I like, when I like.
I see a real upside for pros in getting MF and D800's, distancing themselves from the hordes of wannabees and hobbyists who could afford a few low cost strobes. The added skill required in high resolution anything, from image, moving or still, sound, quarterbacking, writing etc, separates the skilled from those who just have an expensive hobby.
More care will be taken with lighting, scene optimization, makeup, clothing detail, poses...all things that take a talent and attention to detail that few people have the patience for or the understanding of how important they are.
This is nothing new - People shoot MF digital for portrait/fashion every day 40,65 even 80 mp digital backs...whats the big deal ? get a good make up artist, have good technique and take an picture. If it bothers you dont pixel peep at 100%, shoot in DX mode or downsample the file. Retouching a Zit is just as easy on a 36mp file as a 12 mp file !
After dealing with the fashion industry, I would rather have a great MA on my crew than a great camera. A makeup artist who knows photography and which chemicals react to which light sources, and perception. With subtle shading a face can be transformed in person or on screen from interesting to gorgeous, from bland to hard edge punk with a few well placed brush and sponge strokes. What to up your portrait or commercial catalog business, align yourself with a great MA. There is a big difference between a makeup specialist and one who is talented with photo subjects. Stan St Petersburg Russia
>I can barely afford a D800.........let alone a makeup artist! > I don't need a makeup artist with my D700.
You dont "Need" a make up artist with the 800 anymore than you do with the 700. My point was, If you are going to pixel peep a 36mp file and it bothers you there is "too much" detail, downsample to 12MP or shoot DX mode or If you are intent on pixel peeping a 36MP file get a good make up artist, to take care of bad skin, thin eyelashes, wrinkles etc. I know plenty of pros shooting a D700 who use make-up artist - if you are doing a serious shoot it make a huge difference. And the cost is built into the shoot anyway.
"I see a real upside for pros in getting MF and D800's, distancing themselves from the hordes of wannabees and hobbyists who could afford a few low cost strobes. The added skill required in high resolution anything, from image, moving or still, sound, quarterbacking, writing etc, separates the skilled from those who just have an expensive hobby.
More care will be taken with lighting, scene optimization, makeup, clothing detail, poses...all things that take a talent and attention to detail that few people have the patience for or the understanding of how important they are."
Agree with you 100 percent Stan. Could not have said it better. спасибо - spaciba!
Speaking in general to responders on this thread (lot of good responses embedded), this does remind me of the effect of purchasing a great new lens:
- for example when I got my 500VR I spent intensive amounts of time learning and refining my sharpening techniques (still not done, can always get better ).
- my reasoning was that I had better "up my game" in post-processing skills commensurate to what this fine lens can produce. I didn't want to be a great equipment user and a hack at PP.
Same diff with the D800. This is a camera that will force oneself to grow in shooting technique and post-processing knowledge. So if you are getting a D800, now is probably the time to invest in that software package you have been eyeing and/or books and courses on how to use it at an advanced level. Personally even though I've probably advanced a long way from where I started, I still have a lot to learn about Capture NX2 and can always improve more.
All of this "forced to learn" type action is a huge benefit to the shooter for the rest of your shooting career. D800 will be a great camera to learn and grow with IMHO!
Something to keep in mind in assessing the camera's megapixel count is the increase in printer resolution.
The newest higher end printers produce more detail on paper than just a couple of years ago. However, the image must have that detail to reproduce. Conceptions of just three or four years ago as to how many pixels per inch are no longer a satisfactory measure.
Sat 25-Feb-12 03:34 PM | edited Sat 25-Feb-12 03:36 PM by voyageurfred
"Something to keep in mind in assessing the camera's megapixel count is the increase in printer resolution."
You bring up a good point Ralph. Some digital colour printers like Epson, have a suggested resolution setting of 360ppi, for best results, not the 300dpi widely used for off-set presses in the printing industry.
Another item to consider is, because of computer algorithms, how you set your resolution is very important to achieve best results, especially in fine art printing. The setting you use to output your files should be divisible by 8, due to the structure of the binary code and byte order.
If you haven't noticed, the 360ppi for Epson Printers, 240ppi of most Nikon DSLR files, and even the 72ppi resolution of the better computer screens... are all divisible by 8.
Lightroom apparently has advantages over PS in doing the resampling since it can automatically mix different modes of doing so.
So, as to 36 megapixels, I guess the message is to save the files. They may not be excess pixels in the the future and may print or otherwise be processed even better in the future than they can be now.